Market making is the process whereby a market participant, called a market maker, simultaneously and repeatedly posts limit orders on both sides of the limit order book of a security in order to both provide liquidity and generate profit. Optimal market making entails dynamic adjustment of bid and ask prices in response to the market maker’s current inventory level and market conditions with the goal of maximizing a risk-adjusted return measure. This problem is naturally framed as a Markov decision process, a discrete-time stochastic (inventory) control process. Reinforcement learning, a class of techniques based on learning from observations and used for solving Markov decision processes, lends itself particularly well to it. Recent years have seen a very strong uptick in the popularity of such techniques in the field, fueled in part by a series of successes of deep reinforcement learning in other domains. The primary goal of this paper is to provide a comprehensive and up-to-date overview of the current state-of-the-art applications of (deep) reinforcement learning focused on optimal market making. The analysis indicated that reinforcement learning techniques provide superior performance in terms of the risk-adjusted return over more standard market making strategies, typically derived from analytical models.

Do macroeconomic factors matter for stock returns? Evidence from estimating a multifactor model on the Croatian market: Factor models observe the sensitivity of an asset return as a function of one or more factors. This paper analyzes returns on fourteen stocks of the Croatian capital market in the period from January 2004 to October 2009 using inflation, industrial production, interest rates, market index and oil prices as factors. Both the direction and strength of the relation between the change in factors and returns are investigated. The analyses included fourteen stocks and their sensitivities to factors were estimated. The results show that the market index has the largest statistical significance for all stocks and a positive relation to returns. Interest rates, oil prices and industrial production also marked a positive relation to returns, while inflation had a negative influence. Furthermore, cross-sectional regression with the estimated sensitivities used as independent variables and returns in each month as dependent variables is performed. This analysis resulted in time series of risk premiums for each factor. The most important factor affecting stock prices proved to be the market index, which had a positive risk premium. A statistically significant factor in 2004 and 2008 was also inflation, marking a negative risk premium in 2004 and a positive one in 2008. The remaining three factors have not shown as significant.

This article studies house price developments in six European countries: Croatia, Estonia, Poland, Ireland, Spain and the United Kingdom. The main goal is to explore the factors driving the rise of house prices in transition countries. Because house price increases in the last two decades are not peculiar to transition countries, the analysis is extended to three EU-15 countries that have recorded house price rises. The similarities and differences between the two groups of countries in terms of house price determinants can thus be explored. In the first part of the empirical analysis VAR is employed to detect how GDP, housing loans, interest rates and construction contribute to real house price variance. In the second part of the analysis multiple regression models are estimated. The results of both methods suggest that the driving forces behind house price inflation in both groups of countries are very similar and encompass the combined influence of house price persistence, income and interest rates.

We introduce a variant of the Barndorff-Nielsen and Shephard stochastic volatility model where the non-Gaussian Ornstein–Uhlenbeck process describes some measure of trading intensity like trading volume or number of trades instead of unobservable instantaneous variance. We develop an explicit estimator based on martingale estimating functions in a bivariate model that is not a diffusion, but admits jumps. It is assumed that both the quantities are observed on a discrete grid of fixed width, and the observation horizon tends to infinity. We show that the estimator is consistent and asymptotically normal and give explicit expressions of the asymptotic covariance matrix. Our method is illustrated by a finite sample experiment and a statistical analysis of IBM™ stock from the New York Stock Exchange and Microsoft Corporation™ stock from Nasdaq during a history of five years.

This paper investigates the exchange rate pass-through (ERPT) effect in Croatia using a threshold model. To date, empirical studies have failed to find a high degree of ERPT in Croatia, even though exchange rate stability has been one of the goals of monetary policy throughout the transition period. Using a nonlinear model, this study demonstrates that ERPT is present in the data and shows why it cannot be detected with a linear model. According to our estimates, there is a threshold at 5.91 percent of the monthly growth rate of the nominal exchange rate of the German mark (euro), with the 95 percent interval between about 2.7 and 21.8 percent. Consistent with the theories of sticky prices or pricing to market, ERPT is asymmetric around the threshold: below it, the ERPT effect is weak or statistically insignificant, and above it, the effect is strong and significant.